A Novel Location Prediction Algorithm of Mobile Users For Cellular Networks

Abstract

Predicting the location of a mobile user is one of  the  important  issues  in  mobile  computing  systems. Applications of the location prediction include adjusting the bandwidth of the mobile network, the location based services  (LSB),  smart  handover,  etc.  However,  the applications  require  the  execution  time  of  the  User Mobility  Patterns  Mining  (UMPMining)  algorithm  be instantaneous.  In  this  paper,  we  propose  a  new algorithm named Find_UMP for mining next location of a  mobile  user.  Our  algorithm  includes  two  phase  as follows.  In  the  first  phase  (Find_UMP_1),  we  make  to reduce the complexity of the UMPMining algorithm. In the second phase (Find_UMP_2), we perform to reduce the  number  of  transactions  of  the  paths  database. Results  of  our  experiments  show  that  our  proposed algorithm  outperforms  the  UMPMining  algorithm  in terms of the execution time.